is if the sexes are moving the same way; if they are not, the male data will tell us something different from what the female data tell; male population may be growing faster than female, and in this case it is even more important to take both sexes into account. Examples of these two circumstances-true male and female rates of growth being the same, or being different—are provided by England and Wales 1964 and the United States 1966 respectively. England and Wales for 1964 show an intrinsic rate of 11.3 per 1,000 population on the female side, and 11.1 on the male side. One is not tempted to search for the meaning of this difference which may be supposed analogous to sampling variation. The United States figures for 1966 are a different case: female intrinsic increase was 9.7 per 1,000, and male was 12.8. For females the intrinsic rate is calculated as ln -^o In 1 289 ------o= \.LOJ T 26.173 for males we have In 1.451 29.026 = 0.0128, where RQ is the Net Reproduction Rate and T the length of generation. The U.S. difference of 3 per 1,000 population showed separately in white and nonwhite for 1966. It has built up gradually during the 1960's; early in the decade women had the slightly higher rate, but by 1964 females were 15.7 and males were 17.5. Canada for 1966 shows a difference in the same direction as the United States and about half the amount; its female rate was 10.1 and its male 11.5. Such differences are remarkable, since after all the same babies are bom to mothers as to fathers, and the sex ratio of the births is nearly the same in all groups. Why does referring them to fathers make a much higher rate of increase in the population? A possible answer that first comes to mind is that recent fathers may be younger. If they are younger, then the same children referred to fathers would imply a shorter generation, and hence a faster turnover, and this would make a higher intrinsic rate. To check the point we look up the mean ages of mothers and fathers at childbirth in the stationary population, designated //. We find for ju in the male and female calculations respectively for the United States: Male Female 1959-61 29.55 26.37 1964 29.62 26.53 1966 29.33 26.35 diagram may help us follow the preceding arguments.